CN113962915B - Self-adaptive nonlinear super-dynamic image synthesis method under non-uniform illumination condition - Google Patents
Self-adaptive nonlinear super-dynamic image synthesis method under non-uniform illumination condition Download PDFInfo
- Publication number
- CN113962915B CN113962915B CN202111221433.6A CN202111221433A CN113962915B CN 113962915 B CN113962915 B CN 113962915B CN 202111221433 A CN202111221433 A CN 202111221433A CN 113962915 B CN113962915 B CN 113962915B
- Authority
- CN
- China
- Prior art keywords
- image
- exposure time
- saturation point
- neural network
- gray
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000005286 illumination Methods 0.000 title claims abstract description 46
- 238000001308 synthesis method Methods 0.000 title abstract description 8
- 238000013528 artificial neural network Methods 0.000 claims abstract description 30
- 230000003287 optical effect Effects 0.000 claims abstract description 27
- 238000000034 method Methods 0.000 claims description 13
- 230000003044 adaptive effect Effects 0.000 claims description 11
- 238000003384 imaging method Methods 0.000 claims description 10
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 238000003786 synthesis reaction Methods 0.000 claims description 7
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 229920006395 saturated elastomer Polymers 0.000 abstract description 11
- 238000001514 detection method Methods 0.000 abstract description 5
- 238000004364 calculation method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000012417 linear regression Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20172—Image enhancement details
- G06T2207/20208—High dynamic range [HDR] image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Studio Devices (AREA)
Abstract
The self-adaptive nonlinear hyper-dynamic image synthesis method under the non-uniform illumination condition solves the problem that the detected target image is saturated when the image is formed under the non-uniform dark field illumination condition, and belongs to the technical field of optical element detection. Under the condition that the illumination condition is not changed, the illumination field of the area where the damage point is located is not uniform, and n images which change along with the exposure time of the camera are acquired. And (3) with the increase of the exposure time, the gray value of the image of the damage point is correspondingly increased, whether a saturation point appears in the image is judged, if the saturation point appears, the gray value and the exposure time of the corresponding position in the image of the exposure time before the saturation point are utilized to train the neural network, nonlinear regression is realized, the gray value of the exposure time after the corresponding saturation point is obtained by utilizing the neural network, the saturated gray value is replaced, and the ultra-dynamic image is generated.
Description
Technical Field
The invention belongs to the technical field of optical element detection.
Background
Under high power conditions, photo-induced damage to optical elements becomes a troublesome problem that people must solve. In order to detect and track the growth process of the damage, the damage condition of the optical element needs to be detected on line, and the detection principle is shown in fig. 1. With total internal reflection illumination, a damaged part leaks strong light, and a damaged part leaks a small amount of light due to internal reflection, so a dark field image is obtained in which the damaged part becomes a dark background in the image and the damaged part becomes a bright foreground object, as shown in fig. 2.
In order to detect a micro damage point of hundreds of microns, a photosensitive element of an imaging system adopts high-sensitivity parameters, and because the aperture of an optical element used in a high-power scene is usually large and reaches 400mm × 400mm, the distribution of an illumination field is difficult to be uniform, which brings a problem: while the detection capability of the tiny damage is improved, the pixel value of the corresponding image part is often saturated for a larger damage point part, which has adverse effect on the subsequent judgment of the damage size value.
Disclosure of Invention
Aiming at the problem that the detected target image is saturated when the imaging is carried out under the condition of the non-uniform dark field illumination, the invention provides a self-adaptive nonlinear super-dynamic image synthesis method under the condition of the non-uniform illumination.
The invention discloses a self-adaptive nonlinear super-dynamic image synthesis method under a non-uniform illumination condition, which comprises the following steps:
s1, setting the exposure time of the camera to t under the fixed illumination intensityiI is 1, …, n, imaging the tested optical element to obtain n images, and acquiring the optical element image at each exposure time as an image Ii;tnRepresenting a current exposure time;
s2, where the initial value of I is 1, sequentially judges the image I until I becomes n according to the change of timeiIf a saturation point exists, the following is executed:
when i is k, the exposure time t is all before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1};
Using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1And taking the trained neural network as a nonlinear regression function of the corresponding saturation point to obtain the exposure time tnThen, corresponding to the gray regression value of the saturation point;
finding out all saturation points in the n images and nonlinear regression functions corresponding to each saturation point according to S2;
s3, at the current exposure time tnNext, obtaining the gray value of each saturation point; obtaining the exposure time t by using the nonlinear regression function corresponding to each saturation pointnReplacing the corresponding gray value with the corresponding gray regression value to generate the current exposure time tnThe super-dynamic image of (1).
The invention also provides a self-adaptive nonlinear super-dynamic image synthesis method under the non-uniform illumination condition, which comprises the following steps:
s1, in solidSetting the exposure time of the camera to t under the fixed illumination intensityiI is 1,2, …, and the image of the measured optical element collected at each exposure time is image IiThe initial value of i is 1;
s2, collecting the optical element to be tested to image when reaching the exposure time along with the change of time, and judging the current image IiIf the saturation point does not exist, S2 is repeatedly executed, if the saturation point exists, i is equal to k, it is determined whether the saturation point has a corresponding nonlinear regression function, if so, S3 is executed, and if not, the following is executed:
all exposure times t before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1};
Using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1The trained neural network is used as a nonlinear regression function of a corresponding saturation point;
s3, acquiring a gray value of a saturation point; obtaining exposure time t by utilizing nonlinear regression function corresponding to saturation pointkCorresponding gray regression value is used for replacing image IkGray values corresponding to the intermediate saturation points to generate the current exposure time tkThe next super moving picture is transferred to S2.
Preferably, the neural network is a DNN network.
Preferably, the neural network is an MLP neural network of a 4-layer perceptron.
Preferably, in S1, laser light with a fixed light intensity is applied to the optical element side surface.
The method has the beneficial effects that the method is a self-adaptive high-dynamic image synthesis method, and can improve the dynamic range of the image, so that the influence of exposure saturation on the judgment of the size value of the damaged point is avoided.
Drawings
FIG. 1 is a schematic diagram of optical element dark field imaging detection of surface damage based on laser side illumination;
FIG. 2 is an image of an optical element having a damaged surface;
FIG. 3 is a graph of the illuminance unevenness of A, B, C three damage points under the condition of unchanged lighting conditions, and acquiring n images of the exposure time variation of the camera;
fig. 4 is a graph of A, B, C increase in image gray scale values for three damage points as exposure time increases.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments may be combined with each other without conflict.
The invention is further described with reference to the following drawings and specific examples, which are not intended to be limiting.
In order to solve the problem that the size value judgment of the damaged point of the optical element is influenced by the imaging saturation of the high-sensitivity camera under the condition of non-uniform dark field illumination, a self-adaptive high-dynamic image synthesis method is provided, and the dynamic range of an image can be improved, so that the influence of exposure saturation on the size value judgment of the damaged point is avoided.
The technical principle is as follows: under the condition that the illumination condition is unchanged, the illumination field of the area where A, B, C three damage points are located is not uniform, and n images with the exposure time t of the camera changed are acquired. As shown in fig. 3.
As the exposure time t increases, A, B, C the gray scale values of the images of the three damaged points will increase accordingly, and if not saturated, the diagram is shown in fig. 4. Assuming that the illumination field is not uniform and the illumination values a < B < C, a low-to-high distribution of the gray scale curve at A, B, C points occurs.
As the exposure time increased, point C first became saturated, B, A times. And judging whether a saturation point appears in the image, if so, not increasing the gray value along with the increase of the exposure time, as shown in figure 4.
Designing a DNN network (multilayer perceptron MLP) for the value of the unsaturated point at the point C, performing nonlinear regression, and after the regression, corresponding to the saturated section, forming a virtual line segment on the saturated section.
The reason why the nonlinear regression is adopted is that the microstructure of the damage point is very complicated, and the scattering of the illumination light is different for each point. The method herein thus designs an adaptive non-linear regression of the saturation point.
Saturation point regression is performed on a 4K image, and the calculation amount of points needing regression is huge. If the point C is saturated, tens or even hundreds of points around the point C are likely to be saturated, because the algorithm is to perform regression point by point, and the calculation amount is increased greatly. In order to solve this problem, it is proposed to perform regression using the value before saturation only for the point at which saturation occurs, which can greatly reduce the amount of computation and complexity.
And (5) obtaining the value of the virtual line segment in the graph 4 through regression, correcting the saturated value by adopting the value of the virtual line segment, and obtaining the ultra-dynamic range image.
In accordance with the above principle, the present embodiment provides two embodiments, that is, the application scenario of embodiment 1 is at the current exposure time tnAnd generating the current exposure time t on the basis of acquiring n images in totalnIn the application scenario of embodiment 2, the image is collected in real time along with the exposure time, and the super-dynamic image is generated in real time.
Embodiment 1, a method for synthesizing an adaptive nonlinear super-dynamic image under a non-uniform illumination condition, comprising:
step one, under the fixed illumination intensity, setting the exposure time of a camera as ti,i=1,…,n,Imaging the optical element to be measured to obtain n images, wherein the image of the optical element collected in each exposure time is an image Ii;tnRepresenting a current exposure time;
step two, the initial value of I is 1, and the image I is judged in sequence according to the change of time until I is equal to niIf a saturation point exists, the following is executed:
when i is k, the exposure time t is all before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1};
In order to avoid too few sample points for regression, the increase step delta of the exposure time needs to be limited to guarantee the image sequence before saturation { I1,I2,…,Ik-1The data volume of which is greater than a limit value, i.e. K-1 is required>9。
Using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1And taking the trained neural network as a nonlinear regression function of the corresponding saturation point to obtain the exposure time tnThen, corresponding to the gray regression value of the saturation point;
finding all saturation points in the n images and nonlinear regression functions corresponding to each saturation point according to the second step;
step three, in the current exposure time tnNext, obtaining the gray value of each saturation point; obtaining the exposure time t by using the nonlinear regression function corresponding to each saturation pointnReplacing the corresponding gray value with the corresponding gray regression value to generate the current exposure time tnThe super-dynamic image of (1).
In this embodiment, the neural network is a DNN network or an MLP neural network of a 4-layer perceptron.
In the first step of the present embodiment, laser light with a constant illumination intensity is applied to the side surface of the optical element.
Embodiment 2, a method for synthesizing an adaptive nonlinear super-dynamic image under a non-uniform illumination condition, comprising:
step one, under the fixed illumination intensity, setting the exposure time of a camera as tiI is 1,2, …, and the image of the measured optical element collected at each exposure time is image IiThe initial value of i is 1;
step two, along with the change of time, when reaching the exposure time, collecting the tested optical element for imaging, and judging the current image IiIf the saturation point does not exist, S2 is repeatedly executed, if the saturation point exists, i is equal to k, it is determined whether the saturation point has a corresponding nonlinear regression function, if so, S3 is executed, and if not, the following is executed:
all exposure times t before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1}; in order to avoid too few sample points for regression, the increase step delta of the exposure time needs to be limited to guarantee the image sequence before saturation { I1,I2,…,Ik-1The data volume of which is greater than a limit value, i.e. K-1 is required>9。
Using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1The trained neural network is used as a nonlinear regression function of a corresponding saturation point;
acquiring a gray value of a saturation point; obtaining exposure time t by utilizing nonlinear regression function corresponding to saturation pointkCorresponding gray regression value is used for replacing image IkGray scale corresponding to intermediate saturation pointValue, generating the current exposure time tkAnd (5) switching to the step two for the next super-dynamic image.
Example 3: an adaptive nonlinear hyper-dynamic image synthesis device under the condition of non-uniform illumination, comprising:
the lighting device is used for providing illumination with fixed intensity for the optical element to be detected;
a camera connected with the processor for setting the exposure time to tiI1, …, imaging the optical element to be tested, and obtaining an image I corresponding to each exposure timeiObtaining an image I for each exposure timeiSending to a processor;
a processor for judging the image IiIf yes, judging whether the saturation point has a corresponding nonlinear regression function, if not, executing the following steps: when i is k, the exposure time t is all before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1}; using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1The trained neural network is used as a nonlinear regression function of a corresponding saturation point; if the saturation point has a corresponding nonlinear regression function, obtaining the gray value of the saturation point, and obtaining the exposure time t by using the nonlinear regression function corresponding to the saturation pointkCorresponding gray regression value is used for replacing image IkGray values corresponding to the intermediate saturation points to generate the current exposure time tkThe super-dynamic image of (1).
The processor in this embodiment is connected to the illumination device, and is further configured to set the illumination intensity of the illumination device and the exposure time of the camera.
In this embodiment, the neural network is a DNN network or an MLP neural network of a 4-layer perceptron.
Although the invention herein has been described with reference to particular embodiments, it is to be understood that these embodiments are merely illustrative of the principles and applications of the present invention. It is therefore to be understood that numerous modifications may be made to the illustrative embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that features described in different dependent claims and herein may be combined in ways different from those described in the original claims. It is also to be understood that features described in connection with individual embodiments may be used in other described embodiments.
Claims (9)
1. A method for adaptive nonlinear super-dynamic image synthesis under non-uniform illumination conditions, the method comprising:
s1, setting the exposure time of the camera to t under the fixed illumination intensityiI is 1, …, n, imaging the tested optical element to obtain n images, and acquiring the optical element image at each exposure time as an image Ii;tnRepresenting a current exposure time;
s2, where the initial value of I is 1, sequentially judges the image I until I becomes n according to the change of timeiIf a saturation point exists, the following is executed:
when i is k, the exposure time t is all before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1};
Using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1Will finish trainingThe resultant neural network is used as a nonlinear regression function of the corresponding saturation point to obtain the exposure time tnThen, corresponding to the gray regression value of the saturation point;
finding out all saturation points in the n images and nonlinear regression functions corresponding to each saturation point according to S2;
s3, at the current exposure time tnNext, obtaining the gray value of each saturation point; obtaining the exposure time t by using the nonlinear regression function corresponding to each saturation pointnReplacing the corresponding gray value with the corresponding gray regression value to generate the current exposure time tnThe super-dynamic image of (1).
2. A method for adaptive nonlinear super-dynamic image synthesis under non-uniform illumination conditions, the method comprising:
s1, setting the exposure time of the camera to t under the fixed illumination intensityiI is 1,2, …, and the image of the measured optical element collected at each exposure time is image IiThe initial value of i is 1;
s2, collecting the optical element to be tested to image when reaching the exposure time along with the change of time, and judging the current image IiIf the saturation point does not exist, S2 is repeatedly executed, if the saturation point exists, i is equal to k, it is determined whether the saturation point has a corresponding nonlinear regression function, if so, S3 is executed, and if not, the following is executed:
all exposure times t before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1};
Using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1Will train the completion of the spiritUsing the network as a nonlinear regression function corresponding to the saturation point;
s3, acquiring a gray value of a saturation point; obtaining exposure time t by utilizing nonlinear regression function corresponding to saturation pointkCorresponding gray regression value is used for replacing image IkGray values corresponding to the intermediate saturation points to generate the current exposure time tkThe next super moving picture is transferred to S2.
3. The method as claimed in claim 1 or 2, wherein the neural network is a DNN network.
4. The method as claimed in claim 1 or 2, wherein the neural network is a 4-layer perceptron MLP neural network.
5. The method for adaptive nonlinear super dynamic image synthesis under non-uniform illumination conditions as recited in claim 1 or 2, wherein in S1, laser illumination at a fixed illumination intensity is applied to the side of the optical element.
6. An adaptive nonlinear super-dynamic image synthesizing apparatus under non-uniform illumination conditions, comprising:
the lighting device is used for providing illumination with fixed intensity for the optical element to be detected;
a camera connected with the processor for setting the exposure time to tiI1, …, imaging the optical element to be tested, and obtaining an image I corresponding to each exposure timeiObtaining an image I for each exposure timeiSending to a processor;
a processor for judging the image IiIf yes, judging whether the saturation point has a corresponding nonlinear regression function, if not, executing the following steps: when i is k, the exposure time t is all before kiCorresponding image IiComposing a sequence of images { I1,I2,…,Ik-1Extracting an image sequence { I }1,I2,…,Ik-1K-1 gray values { V } of the pixel corresponding to the saturation point1,V2,…,Vk-1}; using grey scale value V1,V2,…,Vk-1Training 1 neural network whose input samples are t1,t2,…,tk-1The corresponding output sample is { V at the corresponding saturation point1,V2,…,Vk-1The trained neural network is used as a nonlinear regression function of a corresponding saturation point; if the saturation point has a corresponding nonlinear regression function, obtaining the gray value of the saturation point, and obtaining the exposure time t by using the nonlinear regression function corresponding to the saturation pointkCorresponding gray regression value is used for replacing image IkGray values corresponding to the intermediate saturation points to generate the current exposure time tkThe super-dynamic image of (1).
7. The adaptive nonlinear hyper-dynamic image synthesis apparatus under non-uniform illumination conditions as recited in claim 6, wherein the processor, in connection with the illumination apparatus, is further configured to set an illumination intensity of the illumination apparatus and an exposure time of the camera.
8. The adaptive nonlinear super-dynamic image synthesis apparatus under non-uniform illumination conditions as recited in claim 6, wherein the neural network is a DNN network.
9. The adaptive nonlinear hyper-dynamic image synthesis apparatus under non-uniform illumination conditions as recited in claim 7, wherein the neural network is a 4-layer perceptron MLP neural network.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111221433.6A CN113962915B (en) | 2021-10-20 | 2021-10-20 | Self-adaptive nonlinear super-dynamic image synthesis method under non-uniform illumination condition |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111221433.6A CN113962915B (en) | 2021-10-20 | 2021-10-20 | Self-adaptive nonlinear super-dynamic image synthesis method under non-uniform illumination condition |
Publications (2)
Publication Number | Publication Date |
---|---|
CN113962915A CN113962915A (en) | 2022-01-21 |
CN113962915B true CN113962915B (en) | 2022-04-29 |
Family
ID=79465694
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111221433.6A Active CN113962915B (en) | 2021-10-20 | 2021-10-20 | Self-adaptive nonlinear super-dynamic image synthesis method under non-uniform illumination condition |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113962915B (en) |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1874499A (en) * | 2006-05-12 | 2006-12-06 | 北京理工大学 | High dynamic equipment for reconstructing image in high resolution |
CN103069453A (en) * | 2010-07-05 | 2013-04-24 | 苹果公司 | Operating a device to capture high dynamic range images |
US8885976B1 (en) * | 2013-06-20 | 2014-11-11 | Cyberlink Corp. | Systems and methods for performing image fusion |
CN104881644A (en) * | 2015-05-25 | 2015-09-02 | 华南理工大学 | Face image acquisition method under uneven lighting condition |
CN106017363A (en) * | 2016-08-07 | 2016-10-12 | 哈尔滨工业大学 | High-dynamic-precision large-working-distance self-collimating device and method |
CN106981065A (en) * | 2016-11-02 | 2017-07-25 | 北京信息科技大学 | A kind of image Absolute Central Moment innovatory algorithm based on exposure compensating |
CN108120680A (en) * | 2017-12-19 | 2018-06-05 | 清华大学 | The stray light minimizing technology and device of micro-imaging based on photoelectric characteristic priori |
CN110095472A (en) * | 2019-05-08 | 2019-08-06 | 湖北工业大学 | High reflective metallic surfaces defect inspection method and system based on HDRI |
CN112651911A (en) * | 2020-12-01 | 2021-04-13 | 广东工业大学 | High dynamic range imaging generation method based on polarization image |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110072051B (en) * | 2019-04-09 | 2021-09-03 | Oppo广东移动通信有限公司 | Image processing method and device based on multi-frame images |
-
2021
- 2021-10-20 CN CN202111221433.6A patent/CN113962915B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1874499A (en) * | 2006-05-12 | 2006-12-06 | 北京理工大学 | High dynamic equipment for reconstructing image in high resolution |
CN103069453A (en) * | 2010-07-05 | 2013-04-24 | 苹果公司 | Operating a device to capture high dynamic range images |
US8885976B1 (en) * | 2013-06-20 | 2014-11-11 | Cyberlink Corp. | Systems and methods for performing image fusion |
CN104881644A (en) * | 2015-05-25 | 2015-09-02 | 华南理工大学 | Face image acquisition method under uneven lighting condition |
CN106017363A (en) * | 2016-08-07 | 2016-10-12 | 哈尔滨工业大学 | High-dynamic-precision large-working-distance self-collimating device and method |
CN106981065A (en) * | 2016-11-02 | 2017-07-25 | 北京信息科技大学 | A kind of image Absolute Central Moment innovatory algorithm based on exposure compensating |
CN108120680A (en) * | 2017-12-19 | 2018-06-05 | 清华大学 | The stray light minimizing technology and device of micro-imaging based on photoelectric characteristic priori |
CN110095472A (en) * | 2019-05-08 | 2019-08-06 | 湖北工业大学 | High reflective metallic surfaces defect inspection method and system based on HDRI |
CN112651911A (en) * | 2020-12-01 | 2021-04-13 | 广东工业大学 | High dynamic range imaging generation method based on polarization image |
Non-Patent Citations (5)
Title |
---|
《Adversarial Text-to-Image Synthesis: A Review》;Stanislav Frolov等;《arXiv》;20210131;第1-19页 * |
《CMOS图像传感器量子效率和满阱容量参数测量方法研究》;刘鹏飞;《中国优秀硕士学位论文全文数据库 信息科技辑》;20180715(第07期);第I140-192页 * |
《Final optics damage online inspection in high power laser facility》;魏富鹏 等;《Optoelectronics Letters》;20190701;第15卷(第04期);第306-311页 * |
《光学元件损伤在线检测图像处理技术》;冯博 等;《强激光与粒子束》;20130715;第25卷(第07期);第1697-1700页 * |
《面向道路照明测量的成像***辐射定标研究》;尚颖英;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑》;20210815(第08期);第C034-255页 * |
Also Published As
Publication number | Publication date |
---|---|
CN113962915A (en) | 2022-01-21 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Abdelhamed et al. | A high-quality denoising dataset for smartphone cameras | |
KR101967089B1 (en) | Convergence Neural Network based complete reference image quality evaluation | |
RU2716843C1 (en) | Digital correction of optical system aberrations | |
US11127117B2 (en) | Information processing method, information processing apparatus, and recording medium | |
CN113168670A (en) | Bright spot removal using neural networks | |
CN111292264A (en) | Image high dynamic range reconstruction method based on deep learning | |
CN111062876B (en) | Method and device for correcting model training and image correction and electronic equipment | |
CN111986106B (en) | High-dynamic image reconstruction method based on neural network | |
US10067327B2 (en) | Image acquisition apparatus | |
KR20200084419A (en) | Appratus for generating moire removing model, method for removing moire and imaging device for removing moire | |
JP2020042760A (en) | Information processing method, information processing device, and program | |
Steffens et al. | Cnn based image restoration: Adjusting ill-exposed srgb images in post-processing | |
CN106204466B (en) | A kind of self-adaptive solution method for Fourier lamination micro-imaging technique | |
CN107220945B (en) | Restoration method of multiple degraded extremely blurred image | |
CN106709888B (en) | A kind of high dynamic range images production method based on human vision model | |
CN113962915B (en) | Self-adaptive nonlinear super-dynamic image synthesis method under non-uniform illumination condition | |
Gao et al. | Image enhancement algorithm based on NF-ICM | |
US11443414B2 (en) | Image signal processing | |
WO2023246392A1 (en) | Image acquisition method, apparatus and device, and non-transient computer storage medium | |
JP6232933B2 (en) | Radiation distortion correction apparatus, road environment recognition apparatus, radial distortion correction method and program | |
JP6487008B1 (en) | High resolution imaging device | |
TWI590192B (en) | Adaptive high dynamic range image fusion algorithm | |
JP2011215695A (en) | Traveling object detection device, method, and program | |
Jenkin | Contrast signal to noise ratio | |
Zhao et al. | Analysis of Image Quality Assessment Methods for Aerial Images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |